Robot paths have in general a plurality of path points. These may be taught, in particular manually, i.e. by manually running up to them with the robot and saving them, or also generated automatically, for example from the CAD data of a robot-supported workpiece to be processed or the like.
It may be necessary to modify robot paths of this type. For example, a new obstacle in the working space of the robot, for example an additional shelf, may necessitate bypassing of the obstacle during transport of a workpiece by the robot. Similarly, the workpieces to be processed and/or the process paths may change, in particular locally, due to revisions of the assigned products.
Up until now, the individual path points of the specified robot path must be individually sequentially modified for this purpose according to company-specific practice, for example, by replacing them with newly taught points. A sequential modification of individual path points of this type is not only very expensive and prone to errors, but also often leads to an undesired change of the remaining robot path in the vicinity of the respectively recently modified path point. Therefore, modifications of this type currently generally demand a high level of experience on the part of the user.